Augmented reality (AR) applications attempt to blend virtual objects into real world images in a manner that appears seamless to the user in order to create a compelling AR experience. Existing AR systems typically rely on a combination of standard camera pose calculations and object recognition techniques that tend to be insufficient on their own to provide a convincing fusion of the real and the virtual. Furthermore, existing techniques that merge homogeneous regions together often have difficulty in joining heterogeneous regions of the same object into a single segment. As a result, virtual objects rendered into the camera's video stream can appear at odd, unnatural angles, and contain visible artifacts. The overall effect is generally awkward and displeasing to the user.
Although the following Detailed Description will proceed with reference being made to illustrative embodiments, many alternatives, modifications, and variations thereof will be apparent to those skilled in the art.